Regional convergence and economic development in the EU:
the relation between national growth and regional disparities
within the old and the new member states
Hans Kramar
Vienna University of Technology
Abstract
While European integration has substantially contributed to economic convergence between the
member states of the EU, the diverging development of highly developed metropolitan regions
and lagging rural areas has become a growing challenge especially for the new member states
in Central and Eastern Europe. In this context the paper inquires to which degree the process of
economic restructuring and catching-up in European countries was accompanied by increasing
spatial disparities. Although a lot of research implicitly assumes that economically growing
countries tend to face a trend towards increasing spatial inequalities, the question, whether there
is a direct relation between total economic growth and spatial divergence, has not been
answered sufficiently yet.
The empirical investigation of regional and national GDP data confirms the trend towards
economic convergence on a national scale between 2000 and 2011, mainly caused by the rapid
growth of the most lagging countries. On a regional scale, however, the process of convergence
was much slower and almost came to an end after the beginning of the global economic crisis
in 2008. The reason for these diverging results can be found in the change of disparities within
the countries: While regional inequalities largely remained unchanged in the majority of the old
member states, the gap between rich and poor regions widened in most countries which
accessed the EU since 2004. This trend, however, slowed down or even reversed after 2008,
which seems to confirm the assumption that economic growth intensifies spatial divergence. A
detailed analysis of the correlation between national growth rates and the change of regional
disparities, however, indicates that growing divergence in the new member states can hardly be
explained by the speed of total economic growth, but rather has to be attributed to other specific
conditions there. A reflection on the mechanisms of agglomeration economies suggests three
arguments for the strong diverging effect of the catching-up processes in the new member
states, which await to be tested empirically in future research.
Keywords: economic growth, regional disparities, convergence, agglomeration economies
JEL code: R12
1. Introduction: Economic growth and regional convergence in literature
In regional science and economic geography the debate on the change of regional disparities
and on the process of spatial convergence has always moved between two poles: Under the
neoclassical paradigm the free play of market forces guarantees the compensation of
inequalities in space as a consequence of mobile labour and capital (Richardson 1973) or trade
of specialized goods (Ohlin 1933). On the contrary, the approach of regional polarization, which
basically goes back to Myrdal’s concept of circular cumulative causation (1957), argues that
spatial inequalities tend to increase in recursive and self-reinforcing processes. While Myrdal
claims that the outflow of labor, capital and resources from the less developed regions
(“backwash effects”) outperform the positive “spread effects” of economic development in
growing regions, Hirschman’s model of “unbalanced growth” (1958) argues that negative
“polarization effects” produce a temporary increase of inequalities, which tends to be
compensated by “trickling-down-effects” and “counter-balancing forces” (e.g. political
intervention) in the long run. Schmidt (1966) substantially contributes to a spatially
differentiated theory of economic growth analyzing the diffusion of investment effects in space.
He argues that the localization of complementary investment effects on other sectors depends
on the stage of development of a country claiming that complementary effects in less developed
countries tend to be strongly concentrated on the location of investment. Similar to Hirschman,
however, Schmidt also expects economic and political counterforces to compensate the short-
term agglomerative impact of investment.
The phenomenon of polarized growth was largely neglected in neoclassical theory until the
“New Economic Geography” implemented agglomeration economies into the neoclassical
model (Krugman 1991, Venables 1996, Puga 1999). This approach is widely considered as an
important step away from the equilibrium principle in economics allowing the explanation of
spatial inequalities and unbalanced growth. In Economic Geography and Regional Science,
however, the investigation of agglomeration economies has a long tradition, which goes back
to the basic works of Weber (1909), Marshall (1920), Hoover (1937) or Isard (1956). Since
then, a lot of theoretical and empirical work has been done to explore the driving forces of
economic concentration in space. In this context the term “agglomeration economies” is not
confined to pecuniary externalities (which, for instance, appear in reduced transport and trade
costs for firms), but covers all kinds of advantages and disadvantages, which result from the
spatial concentration of economic activities. Authors, who explore the mechanisms of
“innovative milieus”, stress the importance of human relations and informal networks in a
limited geographical area (Camagni 1991, Aydalot and Keeble 1988, Maillat 1995) as the main
reason for innovative firms to locate close to related companies. Research on “industrial
districts” strongly emphasizes the importance of “collective efficiencies” (Schmitz 1995) in
localized industrial clusters (Becanttini 1990, Priore and Sabel 1989, Markusen 1996, Harrison
1992), mainly picking up the concept of Marshallian externalities, which regards knowledge-
spillovers, labour market pooling und input sharing as the main agglomeration factors of
industries. Porter (1990) emphasizes the importance of local competition, which tends to
promote knowledge externalities and therefore accelerates the pursuit and adoption of
innovation. Other authors attribute the agglomeration of innovation and growth mainly to
localized knowledge spillovers in industrial clusters (Jaffe 1989, Feldman 1994, Feldman and
Florida 1994). Most of these approaches agree on the fact that the spatial concentration of
economic activities is a necessary, but not a sufficient condition for the emergence of
agglomeration economies. Consequently, there is an extensive debate in literature on the
specific local conditions, which foster or impede agglomeration economies in different spatial
contexts (Breschi and Lissoni 2001, Fritsch 2003) with a strong focus on the role of firm
networks, co-operations and institutional frameworks (Camagni 1994, Cappelin 2003, Bathelt
2003). In this context terms like “embeddeness” (Gravovetter 1985, Oerlemans et al. 2001) or
“untraded interdependencies” (Storper 1997), which refer to the intensity and quality of
relations between economic actors (including informal conventions, habits and rules as well as
mutual trust), have been defined as specific local assets, which promote the agglomeration of
firms. The question, whether these externalities rather appear within a certain branch
(commonly referred to as “localization economies”) or between different branches
(“urbanization economies”) is controversially discussed in respective literature (Glaeser et al.
1992, Audretsch 2003, van der Panne 2004). Despite the heterogeneity of all these approaches,
they all argue for the advantages of spatial concentration, which are responsible for higher
returns of public and private investment in urban centers than in sparsely populated areas. From
that point of view the concept of agglomeration economies suggests that it is not economic
development that produces spatial inequalities, but rather the other way round: Assuming the
existence of positive agglomeration economies, it is the spatial concentration of economic
activity which determines economic growth.
In the broad scientific debate on the mechanisms behind regional convergence or divergence
the influence of national economic growth only plays a minor role. Nevertheless, it was already
in the 1950ies, when Kuznets (1955) postulated that economic development comes along with
a temporary increase of social and spatial inequalities. The Kuznets curve, which says that
growing income per capita first induces rising and then falling disparities, is based on the
assumption that industrialization induces a temporary migration from the rural areas to the cities
due to higher wages. In a comprehensive empirical study, which demonstrates that spatial
inequalities are higher and increase faster in less developed countries over a long period of time,
Williamson (1965) comes to rather similar conclusions: “[…] experience suggests that
increasing regional inequality is generated during the early development stages, while mature
growth has produced regional convergence or a reduction in differentials” (p.44).,
In a more recent study Petrakos et al. (2005) indicate that disparities tend to increase in growth
periods and decrease in times of stagnation or recession. Postulating a procyclical behavior of
economic disparities the authors conclude that “[…] no matter what other factors may affect
the evolution of disparities, economic growth will always generate new imbalances” (p. 1853).
Barrios and Strobl (2009) attribute the causal relation between economic growth and increasing
spatial inequalities to the tendency of knowledge and innovation to agglomerate in space: “[…]
although knowledge and technical progress are in this regard seen as the main engines of
economic growth in the long run, the latter may inevitably increase rather than decrease regional
inequalities, and these two elements are very unlikely to be evenly spread” (p.575). Since the
authors assume that technological and structural changes do not appear in all regions at the
same time, they expect rising regional inequalities “at least at the beginning of periods of high
national growth” (p.582).
In brief, there is a lot of research on polarized growth, which implicitly assumes that economic
development is commonly connected with growing regional disparities. Nevertheless, there is
a lack of explicit empirical evidence whether the speed of total economic growth has a direct
impact on the change of spatial inequalities. Therefore, the relation between national growth
and intranational disparities in the EU member states is empirically analyzed in section 3.
2. Regional and national convergence in the European Union
Recent empirical studies on the change of disparities within the European Union show that
interregional convergence within the member states clearly lags behind international
convergence between them, which especially applies to the new member states in Central and
Eastern Europe (Kramar 2006, Brakman and Marrewijk 2013, Monasteriotis 2014, European
Union 2014). In the first part of the empirical work these findings are verified and updated for
the EU28 between 2000 and 2011. To get a detailed picture of the change of spatial inequalities
in the European Union over that period, the analysis distinguishes between:
2 spatial levels
national level (NUTS-0)
regional level (NUTS-3)
2 groups of countries
old member states (EU15)
accession countries of 2004, 2007 and 2013 (EU13)
3 periods of time
before the enlargement of the EU in 2004 (2000 - 2004)
before the outbreak of the global economic crisis (2004 - 2008)
during the crisis (2008 - 2011)
The economic output is measured by the Gross Domestic Product (GDP) at current market
prices in Purchasing Power Standard (PPS) per inhabitant as provided by Eurostat in April
2015. Economic disparities are expressed by the coefficient of variation of the single (regional
or national) GDP levels. Since the coefficient of variation is a standardized measure of
dispersion, which is defined as the ratio of the standard deviation to the mean of all values, the
results of samples with different ranges can easily be compared. The standard deviation
provides “absolute” disparities, which reflect the differences from the mean value, while the
coefficient of variation indicates “relative” disparities, which are related to the quotients of
regional and national values. This difference is especially important for the comparison of
deviations between different countries, but also for the interpretation of changes over time,
which can be illustrated by a simple example: If all regional GDP per capita values increased
by the same growth rate (e.g. +5%), the standard deviation would rise, while the coefficient of
variation would remain unchanged. If the GDP per capita grew by a constant amount in all
regions (e.g. +200€), the standard deviation would stay the same, while the coefficient of
variation would decrease. For the examination of changing regional inequalities over time the
“relative” view of disparities seems to be the appropriate approach, since equal growth rates in
all regions are commonly interpreted as an unchanged spatial distribution.
The left diagram in Figure 1 confirms that the disparities between the 27 of the 28 current
member states of the EU1 strongly decreased between 2000 and 2011. This development can
be interpreted as a typical process of Beta-convergence (see Barro 1991), which is characterized
by a dynamic development of the poorest countries with growth rates clearly above the
1 Luxembourg is excluded from the analysis due to the special situation of the country, which causes
extraordinary GDP levels
European average.2 Evidently, this trend, which is not influenced by the actual accession of the
new member states in 2004 and 2007, was harshly stopped in the year 2008 when the global
economic crisis reached Europe and threatened economic development there. Economic
stagnation in most countries also had an impact on the process of convergence, since the
catching-up process of the less developed countries was interrupted.
Figure 1: Change of economic disparities in the EU28 (differentiated for “old” and “new”
member states) on the national and on the regional level 2000-2011
Source: Eurostat, own calculations
The separate consideration of the development in the “old” EU15 und the “new” EU13 shows
rather different results: The disparities between the old member states remain at a rather low
level over a long period. A relatively rapid rise of the curve since 2009, however, might indicate
that the crisis affects the lagging countries in Southern Europe most. Differences in national
GDP are much higher between the new member states, but there is a clear trend towards
equalization. This result proves that the process of national convergence within the EU is
mainly caused by a fast growth of the most lagging countries (e.g. Romania, Bulgaria,
Lithuania, Latvia), which were able to reduce the gap not only to the richest countries in
Northern and Western Europe, but also to the well performing new member states in CEE. The
gradient of the function indicates that this trend was not seriously affected by the crisis and
essentially continued over the whole period.
The right diagram in Figure 1, which shows the change of economic disparities between the
European NUTS-3-regions over the same period of time, indicates that inequalities between the
poor and the rich regions are higher in the new member states than in the whole EU. This picture
2 The correlation coefficient between the GDP per capita in the year 2000 and the growth rate in the period 2000
- 2008 is statistically significant (1% level of significance) with a value of -0,714.
reflects the big gap between the metropolitan regions in Central Europe (e.g. Prague, Bratislava,
Budapest, Warsaw) and the rural regions especially in the South-East. Although the deviation
decreased in the period before 2008, the process of convergence was much slower on the
regional level than between the whole states. Furthermore, the economic crisis has reversed the
process since 2008, which is expressed by growing coefficients of variation both for the EU13
and for the whole EU. These numbers clearly reveal that the problem of regional disparities
mainly effects the new member states, while inequalities in the EU15 remained on a much lower
level over the whole period.
To put it in a nutshell, the comparison of national and regional results confirm the initial
assumption that economic convergence in the EU is a success story with regard to the reduction
of development gaps between the whole countries, but not on a regional scale. Additionally, the
ongoing crisis since 2008 has stopped or even reversed the trend towards spatial equality and
therefore aggravated the problem of diverging economic development. For a better
understanding of these processes it seems helpful to go into the matter on an intranational scale
investigating regional convergence within the countries. For that purpose Figure 2 shows the
change of coefficients of variation referring to regional GDP per capita separately for the EU15
and the EU133 countries between 2000 and 2011.
Figure 2: Change of economic disparities within the member states of the EU 2000 - 2011
Source: Eurostat, own calculations
The left diagram in Figure 2 and the numbers in Table 1 indicate that the changes of regional
disparities within most of the old member states were rather moderate between 2000 and 2011.
Only Finland even faced a clear constant trend towards convergence (-2,2%), while only the
3 The graphic does not show Luxemburg, Malta and Cyprus, which consist of less than 5 Nuts-3 regions.
Netherlands (+2,4%) and Ireland (+1,5%) were noticeably confronted by a widening gap
between the rich and the poor regions. Contrary to most of the accession countries (see right
diagram), the poorest countries of the EU 15 (Greece, Portugal, Spain) were not affected by
increasing disparities over the whole period. There are, however, some temporal fluctuations
and short-time peaks in some of the countries: In Greece, France, Ireland and the UK there
period between 2000 and 2004 was characterized by strong regional convergence, which
changed to a process of growing divergence after 2004. In Denmark, Ireland and Sweden the
global economic crisis strongly enhanced spatial inequalities after 2008, while in Belgium,
Austria, Portugal and Finland it rather reduced the economic gap between regions. In spite of
these singular and short-term exceptions, in general the distribution of economic performance
is (and remains) rather balanced in the EU15 countries (+0,1%), with a slight trend towards
growing divergence over the last years.
Table 1: Average annual change of economic disparities within the old member states (EU15)
2000 - 2011 in 3 sub-periods
Country codes sub-periods whole period
2000-2011
number of
Nuts-3-regions 2000-04 2004-08 2008-11
EU15 average -1,2% +0,7% +1,0% +0,1%
AT -1,1% -0,6% -1,6% -1,0% 36
BE +0,1% -1,1% -0,9% -0,6% 45
DE -0,3% -0,7% +0,9% -0,1% 412
DK +0,2% +0,1% +3,6% +1,1% 12
EL -6,6% +4,8% +0,8% -0,6% 51
ES -2,7% -1,5% +1,8% -1,1% 60
FI -1,5% -4,1% -0,6% -2,2% 20
FR -1,3% +2,6% +1,0% +0,7% 101
IE -2,3% +1,5% +6,9% +1,5% 8
IT -1,1% +0,6% +0,5% -0,1% 111
NL +2,9% +3,5% +0,3% +2,4% 41
PT +0,4% +0,1% -1,8% -0,3% 31
SE -2,0% +0,9% +2,9% +0,4% 22
UK -1,7% +3,9% +0,7% +1,0% 140
Source: Eurostat, own calculations
The results in the right diagram and in Table 2 show a totally different picture for the new
member states in CEE. All countries faced growing regional disparities between 2000 and 2008
with very strong increases in Bulgaria, Lithuania, Czech Republic, Romania and Slovenia. A
possible explanation for this striking trend can be found in the increasing competitive pressure,
which forces both economic actors and governments to make use of agglomeration economies
and to concentrate their activities on the few competitive economic centers. A substantial
argument for increasing inequalities in the EU during 1990ies is prepared by Gianetti (2002),
who attributes the diverging development of economically advanced and traditional regions to
growing economic integration, which she expects to foster international knowledge spillovers
at the expense of diffusion processes within the countries. In a similar way, Monfort and
Nicolini (2000) argue that economic integration and liberalization of markets enhances spatial
divergence within countries. Investigating the effects of interregional and international
transaction costs on convergence, they conclude that a reduction of transaction costs tends to
favor the spatial clustering of economic activities: “In this perspective, movements towards
integration and international trade liberalization could be considered as factors possibly
favoring the emergence of regional economic agglomeration inside countries.” (p. 304)
Table 2: Average annual change of economic disparities within the new member states 2000 -
2011 in 3 sub-periods
Country
codes
sub-periods whole period
2000-2011
number of
Nuts-3-regions 2000-04 2004-08 2008-11
EU accession in 2004
average +2,8% +1,2% -0,7% +1,2%
CZ +3,9% +2,5% -1,9% +1,8% 14
EE +5,5% -1,6% +0,7% +1,6% 5
HU +1,0% +3,0% +1,3% +1,8% 20
LT +5,1% +1,9% -2,2% +1,9% 10
LV +1,5% +0,4% -7,9% -1,6% 6
PL -0,1% +1,0% +1,7% +0,8% 66
SI +4,2% +1,2% -0,8% +1,7% 12
SK +1,3% +1,1% +3,5% +1,8% 8
EU accession in 2007
average +0,9% +8,6% +1,3% +3,7%
BG +2,6% +10,4% +1,9% +5,1% 28
RO -0,7% +6,8% +0,7% +2,4% 42
EU accession in 2013
HR +2,9% -0,9% +2,1% +1,3% 21
Source: Eurostat, own calculations
Surprisingly, this argumentation cannot applied to explain to the situation in most of the
accession countries of the year 2004, where the speed of divergence rather slowed down after
formal accession. Only Poland and Hungary had a slight acceleration of divergence after their
accession in 2004, starting from a comparatively low increase of disparities between 2000 and
2004. The comparison with the development in the EU15 countries reveals that the change of
disparities after the EU-integration does not significantly deviate anymore from the values of
the old-established EU-members in the same period. The rapid increase of regional disparities
between 2004 and 2008 in Bulgaria and Romania suggests that divergence tends to speed up
during the accession process the just before the factual EU-integration.
The total reversal of this trend happened after the year 2008, when decelerated growth retarded
the increase of regional disparities or even reduced them in countries like Latvia, Lithuania,
Czech Republic and Slovenia. From the present state of available data it is not possible to
predict whether this trend has continued until today and how it will proceed in future. The
results, however, suggest that accelerating growth in the CEE countries would probably bring
the problem of growing disparities back on stage.
3. The relation between national growth and intranational disparities
The results presented in the previous section clearly indicate that the catching-up process of the
new member states of the EU is accompanied by growing regional divergence between
booming urban centres and lagging rural areas, while spatial disparities in most of the well-
established states in Western and Central Europe largely remained unchanged or even
decreased. Furthermore, the sudden slump of economic growth in 2009 and the decelerated
development in the following years, were associated with a slowdown of divergence in most of
the new member states. These findings might lead to the hypothesis that the speed of total
economic growth determines the change of disparities within a country.
This assumption is empirically tested by a simple correlation analysis, which opposes the
average annual GDP growth rates of the member states to the average annual change of
disparities within the member states from 2000 to 2011. In spite of the methodological problems
mentioned in section 2, regional disparities are measured both by the standard deviation and by
the coefficient of variation of GDP per capita of NUTS-3-regions within the countries. First,
the relation between total national growth and the change of regional disparities in the member
states of the EU28 is graphically presented in two scatter plots (see Figure 3). Then, the
correlation coefficients are shown in a methodological, spatial and temporal differentiation:
Table 3 presents separate coefficients for
2 indicators of regional disparity
standard deviation
coefficient of variation
3 groups of countries (whole EU and 2 sub-groups)
whole EU (EU28)
old member states (EU15)
accession countries of 2004, 2007 and 2013 (EU13)4
4 periods of time (whole period and 3 sub-periods)
whole period (2000 - 2011)
before the enlargement of the EU in 2004 (2000 - 2004)
before the outbreak of the global economic crisis (2004 - 2008)
during the crisis (2008 - 2011)
Figure 3: Correlation between total national growth and the change of regional disparities
(annual average 2000-2011)
Source: Eurostat, own calculations
The left diagram in Figure 3 provides a striking picture for the relation between the average
annual national growth rate and the change of regional disparities measured by the standard
deviation of GDP per capita in the Nuts-3-regions of a country. The obviously positive
correlation in the scatterplot is confirmed by a highly significant (at the 0.01 level) correlation
coefficient (+0,918), which is also significant for the two periods before the crisis, but not for
the time after 2008 (see Table 3). These results clearly indicate that the absolute deviations of
regional GDP per capita from the national average rise with the national growth rates of a
country, which can, however, be partly explained by a statistical size effect. In this approach
constant regional differences from a growing mean of the sample (national GDP) would provide
unchanged disparity levels, whereas in a relative consideration they would rather be interpreted
as a sign of convergence.5
4 Again, the graphic does not include Luxemburg, Malta and Cyprus, which consist of less than 5 Nuts-3 regions. 5 See discussion of both measures of dispersion for the description of regional disparities in section 2.
In order to eliminate the size effect caused by national growth and to consider relative deviations
from national GDP levels, a second correlation analysis uses the coefficient of variation instead.
The right diagram, which presents the relation between average annual growth of GDP and the
average annual change of the coefficient of variation of regional (Nuts-3) GDP per capita values
between 2000 and 2011, provides a less impressive but still a distinct result: In spite of higher
unexplained deviations the clearly positive correlation coefficient (+0,462) is still sufficiently
significant at the 0.05 level. Again, the first two sub-periods also provide positive (but only
party significant) results, while there is no evidence for any relation between growth and
changing spatial inequalities after 2008. Nevertheless, the empirical findings for the whole
sample (25 member states of the EU 28) confirm a statistical relation between national
economic growth and increasing relative deviations of regional GDP per capita at least for the
period before the economic crisis.
Table 3: Correlation coefficients between total national growth and the change of regional
disparities in different periods and groups of EU member states
Average annual change of regional disparities on the NUTS-3-level measured by
Standard deviation Coefficient of variation
2000-2011 2000-2004 2004-2008 2008-2011 2000-2011 2000-2004 2004-2008 2008-2011
EU28 +0,918** +0,789** +0,896** +0,333 +0,462* +0,425* +0,314 -0,023
EU15 +0,281 -0,177 +0,146 +0,630* -0,209 -0,731* -0,335 -0,077
EU13 +0,740** +0,539 +0,807** +0,324 +0,106 +0,120 +0,328 +0,199
** significant at the 0.01 level * significant at the 0.05 level
Source: Eurostat, own calculations
The differentiated graphic presentation of the “old“ EU15 countries and the new member states
(“EU13”) in both diagrams, however, disclose two strongly diverging paths of development.
While the EU15 countries show relatively low growth rates and (with some exceptions)
stagnating or decreasing variation coefficients, most of the new member states in CEE face a
catching-up process which is characterized by fast economic growth and growing regional
divergence. Whereas the left diagram indicates that there is still some explaining power of
national GDP growth rates, the right diagram shows no pattern of correlation within the two
groups of countries and therefore suggests that the change of regional economic disparities is
less determined by national growth rates but rather by the type of country considered. Even if
the correlation coefficients separately calculated for the two groups of member states have to
be interpreted with caution due to the small sample6, they basically confirm the graphic
impression: Referring to the change of relative disparities (as measured by the coefficient of
variation) there is no statistical evidence of a significant relation between national growth and
the change of disparities within the two country groups (see Table 3).
Although the coefficients for the new member states (“EU13”) show positive values for all
periods, the deviation of different national development paths seems to be too big to provide
significant results. The totally different ways of convergence in Bulgaria and Latvia, where
similar annual growth rates (+7%) were connected with exploding disparities in the first case
and with spatial equalization in the other, or the special situation of Poland, where prosperous
economic development did not seriously increase regional inequalities, suggest that there must
be other driving forces of spatial divergence. The analysis for the group of old member states
(“EU15”) even provides negative correlation coefficients. Even though the results are not
significant for most periods, they slightly indicate that the faster growing countries of the EU15
tend to overcome their spatial inequalities more easily. The clearly positive correlation results
for the EU13 when using the standard deviation for measuring regional disparities, confirm that
the high growths rates in the new member states cause increasing absolute deviations of regional
GDP per capita from the national average. This result, however, can mainly be attributed to the
statistical size effect in growing economies and not to a relative intensification of economic
inequalities between the Nuts-3-regions of the countries.
4. Conclusions
The empirical investigation of changing spatial disparities in the European Union between 2000
and 2011 indicates that economic convergence between the whole countries was much faster
than on the Nuts-3 level until the year 2008 and almost came to a stop after the begin of the
worldwide economic crisis. The gap between national and regional convergence can mainly be
explained by growing regional inequalities within the majority of the new member states.
Surprisingly, the diverging development of booming urban centers and lagging rural areas
seems to be fastest during the final phase of the accession process and to slow down in most
cases after the formal accession. This phenomenon indicates that the radical conversion of
economic, social and political structures as a pre-condition for EU membership tends to increase
6 Only countries with at least 5 Nuts-3 regions (14 of the EU15 and 11 of the EU13) were included in the
analysis
regional disparities before EU-integration, while the actual accession and the subsequent access
to EU structural funds helps to reduce the economic gap between “rich” and “poor” regions.
The results presented in section 3 suggest that the apparent correlation between the speed of
national economic growth and the intensity of regional divergence can mainly be traced back
to the specific development paths of the EU15 on the one hand and the new member states on
the other. In other words, it seems to be the particular situation in the CEE countries which is
responsible for growing spatial inequalities and not the actual speed of national economic
growth. It can be confirmed that the catching-up process in the new member states goes hand
in hand with increasing disparities between growing economic centers and lagging rural areas,
but there is no clear evidence that the degree of divergence is higher in the fastest growing
countries. The question remains, which of the specific conditions in the new member states are
responsible for the strong disintegrative effect of economic development there. Based on
various arguments from literature on polarized growth and on empirical evidence given in this
paper there are three possible hypotheses, which might make a good case for explaining the
phenomenon of increasing spatial inequalities in catching-up CEE countries:
First, it can be argued that the intensity of agglomeration economies depends on the distribution
of relevant production factors in space. Since the productivity of private investment tends to
increase with capital endowment, infrastructure supply and labour skills in a region, big
regional differences in relevant production conditions promote growing inequalities in GDP. In
other words, agglomeration effects need a minimum amount of basic facilities and amenities to
unfold and to amplify economic activities: “economic growth has a tendency to be associated
with some sort of agglomeration and requires a minimum threshold of resources and activities
in order to take place.” (Petrakos et al. 2005, p.1838). In most of the higher developed countries
regional differences in basic location factors are less pronounced than in the new member states,
where the lagging regions are often poorly equipped with very basic features, which makes
more challenging economic activities almost impossible. From that point of view, the highly
uneven distribution of relevant production factors between well equipped urban centers and
poorly resourced rural regions seems to be a possible reason for growing disparities in the new
member states in CEE.
Secondly, it is well plausible to claim that the emergence of positive trickling-down effects
depends on the degree of integration of the economic system. Since the main handicap of the
new member states can be attributed to the “adverse legacy effects of these earlier non-
democratic governance systems”, which include “the use of outdated technologies, insufficient
updated infrastructure, contaminated land, and institutional and governance systems with
limited capacities and capabilities” (McCann 2015, p.19), they seem to be strongly
disadvantaged in this dimension. Assuming that the quality of built facilities (e.g. cross-linked
infrastructure networks), the interconnectedness of economic actors (e.g. firm networks, co-
operations, linkages)7 and the efficiency of the governance system clearly lag behind the well-
integrated countries in Northern and Western Europe, this line of argumentation suggests that
growing disparities in the new member states are a consequence of less integrated economic
systems, which impede the diffusion of positive spread effects from the booming growth poles
to the lagging regions.
Thirdly, the drastic conversion of economic, social and political structures in the new member
states be held responsible for the widening economic gap between “rich” and “poor” regions.
Economic structural change, social and demographic transformation and the restructuring of
the political system seem to benefit especially established centers, which are able to adopt new
trends and innovations and to take economic advantage from them. On the contrary, these
processes can be a serious threat for less developed regions, which often do not have the
flexibility and the facilities to make use of changing conditions. The strong increase of regional
disparities before the actual EU-integration (see section 2) indicates that this effect mainly
appears in the pre-accession period, when the countries face the most radical changes to
approach EU standards.
The insight that it is not the speed of economic growth, but the particular conditions in the CEE
countries, which are responsible for fast spatial divergence in their catching-up process,
indicates that additional research is necessary in order to assess the driving forces of increasing
spatial disparities. The strongly diverging paths of convergence in different countries would
need additional factors to be explained thoroughly. For that purpose, the three hypotheses given
provide arguments, which await to be tested empirically. The definition of suitable indicators
for (a) the uneven distribution of endowment, (b) the integration of the economic system and
(c) the conversion of economic, social and political structures will definitely be a challenging
task, especially due to the limited availability of suitable data on the regional level. In a
multivariate analysis, which investigates the driving forces of regional divergence within
countries, the national growth rate would then only be one of several explanatory variables.
7 often referred to as “social capital”, “embeddeness”, “institutional thickness” or “untraded interdependencies”
in literature (see section 1)
References
Aydalot, P., Keeble, D. (1988) High Technology Industry and Innovative Environments in
Europe: An Overview. In: Aydalot, P., Keeble, D. (ed.) Technology Industry and Innovative
Environments: The European Experience, London: Routledge.
Barriosa, S., Strobl, E. (2009) The dynamics of regional inequalities. In: Regional Science
and Urban Economics, Volume 39, Issue 5, p.575-591.
Barro, R. J. (1991) Economic growth in a cross section of countries. In: Quarterly Journal of
Economics 106, p.407-443.
Bathelt, H. (2003) Geographies of Production: Growth Regimes in Spatial Perspective 1 -
Innovation, Institutions and Social Systems. In: Progress in Human Geography, Vol. 27,
p.789-804.
Becattini, G. (1990) The Marshallian industrial district as a socio-economic notion. In: Pyke,
F., Becattini, G., Sengenberger, W. (ed.) Industrial districts and inter-firm co-operation in
Italy. International Institute for Labour Studies: Geneva, p.37-51.
Brakman, S., van Marrewijk, C. (2014) Reflections on cluster policies. In: Cambridge Journal
of Regions Economy and Society, 6.2., p.217-231.
Breschi, S., Lissoni, F. (2001) Localised knowledge spillovers vs. Innovative milliex:
Knowledge „tacitness“ reconsidered. In: Papers in Regional Science, Vol.80, Number 3,
p.255-273.
Camagni, R. (1991) Introduction: From the local ‚milieu’ to innovation through cooperation
networks. In: Camagni, R. (ed.): Innovation networks: spatial perspectives, London.
Camagni, R. (1994) Space-time in the concept of „milieu-innovateur“. In: Beiträge zur
Arbeitsmarkt- und Berufsforschung 184, p.74-89.
Cappelin, R. (2003) Networks and Technological Change in Regional Clusters. In: Bröcker,
J., Dohse, D., Soltwedel, R. (Ed.) Innovation Clusters and Interregional Competition, Springer
Verlag: Berlin, Heidelberg, p.52-78.
European Union (2014) Investment for jobs and growth: Promoting development and good
governance in EU regions and cities. 6th Report on Economic, Social and Territorial Cohesion.
Office for Publications of the European Union: Luxembourg.
Feldman, M.P. (1994) The Geography of Inovation. Kluwer Academic Publishers: Dortrecht,
Boston, London.
Feldman, M.P., Florida, R. (1994) The Geographic Sources of Innovation: Technological
Infrastructure and Product Innovation in the United States. In: Annals of the Association of
American Geographers 84, p.210-229.
Fritsch, M. (2002) Measuring the Quality of Regional Innovation Systems: A Knowledge
Production Function Approach. In: International Regional Science Review 25/1, p.86-101.
Gianetti, M. (2002) The effects of integration on regional disparities: Convergence, divergence
or both? In: European Economic Review, Volume 46, Issue 3, p.539-567.
Glaeser, E., Kallal, H., Scheinkman, J., Shleifer, A. (1992) Growth of Cities. In. Journal of
Political Economy 100. p.1126-1156.
Granovetter, M. (1985) Economic Action and Social Structure: The Problem of
Embeddedness. In: American Journal of Sociology 91, p.481-510.
Harrison, B. (1992) Industrial Districts: Old Vine in New Bottles?, Regional Studies 26,
p.469-483.
Hirschman, A.O. (1958) The Strategy of Economic Development. Yale University Press: New
Haven.
Hoover, E.M. (1937) Location Theory and the Shoe and Leather Industries. Havard Economic
Studies 55, Cambridge, Oxford.
Isard, W. (1956) Location and Space Economy. Agegeral theory relating to industrial location,
market areas, land use, trade and urban structure. MIT Press: Cambridge.
Jaffe, A. (1989) Real effects of academic research. In: American Economic Review 79,
p.957-970.
Kramar, H. (2006) Economic Convergence on Different Spatial Levels: The Conflict between
Cohesion and Growth. In: Raumforschung und Raumordnung, 1/2006, Carl Heymanns Verlag:
Bonn, p.18-27
Krugman, P. (1991) Increasing Returns and Economic Geography. In: Journal of Political
Economy 99, p.483-499.
Kuznets, S. (1955) Economic Growth and Income Inequality. In: The American Economic
Review. 45, no.1, p. 1-28.
Maillat, D. (1995) Territorial dynamic, innovative milieus and regional policy. In:
Entrepreneurship and regional development 7, p.157-165.
Markusen, A. (1996) Sticky Places in Slippery Space: A Typology of Industrial Districts. In:
Economic Geography, Vol. 72, No. 2: p.294-314.
Marshall, A. (1920) Principles of Economics. Macmillan: London.
McCann, P. (2015) The Regional and Urban Policy of the European Union: Cohesion, Result-
orientation and Smart Specialization. Edward Elgar Publishing: Cheltenham/UK and
Northampton/MA.
Monasteriotis, V. (2014) Regional Growth and National Development: Transition in Central
and Eastern Europe and the Regional Kuznets Curve in the East and the West. In: Spatial
Economic Analysis 9.2., p.142-161.
Monfort, P., Nicolini, R. (2000) Regional Convergence and International Integration. In:
Journal of Urban Economics, Volume 48, Issue 2, p.286-306.
Mydral, G. (1957) Economic Theory and Underdeveloped Regions. Methuen: London.
Oerlemans, L.A.G., Meeus, M.T.H., Boekema, F.W.M. (2001) Firm clustering and
innovation: Determinants and effects. In: Papers in Regional Science, Vol.80, Number 3,
p.337-356.
Ohlin, B. (1933) Interregional and International Trade. Harvard University Press: Cambridge.
Petrakos, G., Rodriguez-Pose, A., Rovolis, A. (2005) Growth, integration, and regional
disparities in the European Union. In: Environment and Planning A, volume 37, p.1837-1855.
Porter, M. (1990) The Competitive Advantage of Nations. Basic Books: New York.
Priore, M.J., Sabel, C. (1985) Flexible Spezialisierung. In: Piore, M., Sabel, C. (ed.) Das
Ende der Massenproduktion. Wagenbach: Berlin, p. 286-307.
Puga, D. (1999) The Rise and Fall of Regional Inequalities. In: European Economic Review
43, p.303-334.
Richardson (1973) Regional Growth Theory. Wiley: New York.
Schmidt, H. (1966) Räumliche Wirkung der Investitionen im Industrialisierungsprozess:
Analyse des regionalen Wirtschaftswachstums. Opladen: Köln.
Schmitz, H. (1995) Collective Efficiency: Growth Path for Small-Scale Industry. In: Journal
of Development Studies, vol. 31, No. 4, p.529-566.
Storper, M. (1997) The regional world: territorial development in a global economy. Guilford
Press: New York.
van der Panne, G. (2004) Agglomeration externalities: Marshall versus Jacobs. In: Journal of
Evolutionary Economics vol.14. p.393-604.
Venables, A.J. (1996) Equilibrium locations of vertically linked industrries. In: International
Econmic Review 37, p.341-359.
Weber, A. (1909) Über den Standort der Industrie. JBC Mohr: Tübingen.
Williamson, J.G. (1965) Regional Inequality and the Process of National Development: A
Description of the Patterns. In: Economic Development and Cultural Change, Vol. 13, p. 3-45.
Top Related